Australasian Science: Australia's authority on science since 1938

It’s a Wiggly, Wiggly Universe

Image of cosmic microwave background

Figure 1. The Cosmic Microwave Background as revealed by NASA’s WMAP satellite. This is a picture of the whole sky in microwaves, and shows the fluctuations of matter in the Universe only 400,000 years after the Big Bang. The sky is covered in little hot and cold spots of size ~1°, corresponding to the distance sound can travel in the early Universe. Image: NASA / WMAP Science Team

By Karl Glazebrook

A map of the universe as it existed six billion years ago is close to completion, and may provide new insights into the physics of dark energy.

In the Beginning there was Light. But there was also Sound and Fury...

13.7 billion years ago our universe began in the Big Bang, when the whole of infinity was compressed to a singular point. While we do not yet understand the moment of singularity, cosmologists such as Stephen Hawking see that as their ultimate quest.

But putting aside the first 10–32 seconds or so we think we understand the Big Bang pretty well. After all it was simple, almost entirely uniform and devoid of any structure except for tiny fluctuations. The physics of such a medium as it expands and cools can be described almost exactly.

It is from these fluctuations that the giant galaxies we see today grew, and we know our physical model of them is very accurate as we can observe them in fine detail in the Cosmic Microwave Background (Fig. 1). Although our map of the Cosmic Microwave Background looks random, it turns out that the magnitude of the random fluctuations is different on different scales. For example, 1° scales show much bigger fluctuations than larger or smaller scales. What causes these patterns?

The answer lies in the physics of sound, for the universe was not always a vacuum through which sound cannot propagate. For the first few hundred thousand years after the Big Bang it was very thick and dense, and acoustic waves in the density of matter would propagate freely. These were not quite the same as the familiar sound waves in the air around us, where atoms smash into atoms and push each other along. Back then the universe was a coupled “photon–baryon fluid”, which essentially means that it was a plasma where photons would bump into free electrons that would bump into more photons, and protons would just come along for the ride. The net effect would be density waves in the medium propagating at just over half the speed of light.

So what happened to this cauldron? The universe expanded and cooled, and eventually the electrons combined with the protons to make hydrogen atoms. The electrons become trapped in atomic orbits where they could only interact with light in very specific “quantised” ways. The universe finally became transparent for the first time in its existence and the leftover photons, apart from those that hit our telescopes today, essentially then travel forever.

The universe is still transparent today, which is why we can see so far from our backyards, and 99.9999% of those original Big Bang photons are still travelling. The ones we do observe make up our image of the microwave background – a frozen snapshot of the sound waves as they were travelling 400,000 years after the Big Bang, when it became transparent.

This is the meaning of the 1° scale: it corresponds to the distance that sound could travel in 400,000 years, and can be thought of as a fundamental harmonic of the early universe – much like the deepest note of an organ pipe. Like any harmonic there are higher ones and overtones, and these make up the complex pattern we see in the Cosmic Microwave Background. The harmonics and overtones can now be measured with exquisite precision in the Cosmic Microwave Background, revealing a host of cosmological information.

Remarkably, though, we can still see the faint traces of these primordial sound waves in the universe today because the density variations frozen out in the early universe seed and modulate the formation of galaxies. The scale of this today is very large – the sound “wavelength” is about 490 million light years, which is much bigger than any of the galaxy clusters and super-clusters.

This scale imprints itself as a subtle wiggly modulation on the clustering of galaxies called “baryon acoustic oscillations” (Fig. 2).This imprint can only be recovered by statistical analysis of large surveys of more than 100,000 galaxies, they have been detected in the nearby universe – a faint fossil remnant telling us that the seeds of the modern universe were indeed laid down in the Big Bang.

This is a remarkably useful feature of the universe: it has it’s own built-in standard measuring rod woven right into it. Measuring rods are a very useful thing for cosmologists who want to measure the universe at different times. Its geometry and expansion rate are some of our most pressing questions. To determine these we need to have a measuring rod that we can use for distant objects deep in the universe’s past.

For the past decade we have been using Type 1a supernovae to measure the shape and growth of the universe. These exploding stars all have a very similar luminosity. When cosmologists first observed them in the distant universe they found them somewhat dimmer than expected, and this led to the discovery that the expansion of the universe has been accelerating for the past six billion years (AS, Jan/Feb 2009, pp.12–14).

This is now attributed to a new kind of field in the universe called “dark energy”, although such labelling does not mean that it is understood. It’s as mysterious as an apple falling up from a tree and accelerating away into deep space!

Understanding dark energy is one of the key problems in physics and cosmology. If we could measure how fast the universe’s acceleration changes with time it could give us key physical insights.

To do this we need better and alternative measuring rods. We don’t understand exploding stars that well or why they should be so similar, nor quite why they depart from similarity in quite the way they do. There is always the lingering suspicion that maybe they could have exploded a bit differently in the past, which would mean that our ruler would be erroneously calibrated.

But the baryon acoustic oscillation measuring rod comes gold-plated and certified. Its scale can be worked out from the very simple physics of the early universe, and it can be accurately calibrated using the Cosmic Microwave Background.

To use it, “all” you have to do is measure the distances and three-dimensional positions of more than 100,000 galaxies. This must be done for incredibly faint galaxies far away in the universe at the time when the universe started accelerating around six billion years ago. If this can be done accurately, with all the possible sources of error in such a large survey kept under incredibly fine control, one can in principle measure the wiggles. Nothing, of course, is perfect, but any remaining error would be of a completely different kind to those affecting supernovae. Detection of wiggles would give us a new approach to solving the mystery of cosmic acceleration.

In 2006 the WiggleZ Dark Energy Survey set out to do precisely that. While it is straightforward to measure distances using the redshift of galaxies, the problem is to do this for more than 100,000 galaxies over nearly 1000 square degrees of sky.

Fortunately Australian astronomers had access to a fairly unique instrument: the Two Degree Field (2dF) system on the 3.9-metre Anglo-Australian Telescope (AAT) in Coonabarabran, NSW. 2dF is capable of measuring 392 spectra simultaneously. It does this using 392 optical fibres that feed into large spectrographs. An automated robot system positions the fibres at the locations in the 2°-wide focal plane corresponding to those of galaxies, and it can set up the next field while observing the previous field, making for very efficient observing.

2dF had been used once before on this scale: the 2dF Galaxy Redshift Survey made the first 3D map of the southern sky, and catalogued 220,000 galaxies over 2000 square degrees. This survey, taking 200 nights of AAT time, together with the Sloan Digital Sky Survey in the north, revealed the original sound fingerprints in the local universe.

To do this new dark energy survey the WiggleZ team had be able to do the same thing but for galaxies five times further away. This was achieved through a combination of two factors. Firstly, in 2006 the 2dF acquired new and more sensitive spectrographs. Secondly, the WiggleZ survey was able to cherry pick the most star-forming galaxies at these redshifts using new imaging data from NASA’s GALEX ultraviolet space satellite. Galaxies that are bright in the ultraviolet spectrum tend to show strong emission lines in their optical spectrum. Since they light up like a neon sign, a short exposure can secure a redshift and hence a large area of sky can be covered quickly.

This strategy was not without its problems. Every spectrum had to be manually scrutinised to make sure that the right emission lines were picked – a painstaking and tedious job.

The WiggleZ Dark Energy Survey has been a huge observational effort. Every year, 60–70 nights of telescope time had to be manned. Every night 3000 spectra were taken. Piece-by-piece a 3D map of the universe six billion years ago was constructed, validated and analysed statistically for signs of wiggles.

Of course, this survey has been used for more science than merely searching for these faint cosmic traces. The distribution of the luminosity of star-forming galaxies and the evolution of star-formation in the universe has also been measured. Galaxies with exotically high rates of star formation were found and followed up with larger telescopes to determine the mechanisms driving such activity. The clustering of the galaxies has been used to test dark matter-based models for their formation and to look for subtle deviations from Einstein’s theory of gravity.

However, the ultimate question remained outstanding: would the position of the wiggles back then agree with the “standard model” of the accelerating universe? Would there be subtle deviations indicating the new physics of dark energy?

Unfortunately the final chapter in this story is not yet written, but the answer will come some time towards the end of this year. Right now the WiggleZ team is filling in the last holes in their map, a critical requirement to tease out with the greatest possible confidence the subtle cosmic signal.

When that is done the cosmic wiggles, six billion years ago, may paint a new picture of the nature of the accelerating universe.

Karl Glazebrook is Professor of Astrophysics at Swinburne University, and a founding member and joint-leader of the WiggleZ survey (wigglez.swin.edu.au).